Journal of Public Health: From Theory to Practice (2019) 27:581–590 https://doi.org/10.1007/s10389-018-0980-y

ORIGINAL ARTICLE

A survey to validate the traditional perception of diabetes mellitus

Amulya Vijay1 & Priyadharshan Ranganathan1,2 & Balachandar Vellingiri2

Received: 4 April 2018 /Accepted: 12 September 2018 /Published online: 29 September 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract Aim To investigate the influence of gender, marital status, specific food, and lifestyle habits that could possibly be the causative factor according to ancient medical texts. Exploring the alternative medicinal knowledge about diabetes mellitus (DM) and validating its authenticity by carrying out a modern-day study with a scientific perception in identifying the specific food and lifestyle causing DM. Subjects and methods Our study was conducted during the period 2015 to 2017 in Tamil Nadu and Kerala States, . Data was collected from 1206 individuals aged between 18 and 89 years through a predesigned questionnaire which included seven novel questions acquired from the traditional Indian texts of Siddha . Results A questionnaire-based study conducted using which a total of 1206 subjects (men: 55.4%; women: 44.6%) were evaluated with a mean age of 42.88 ± 13.212 years. A higher number of males were affected than females. It is also evident that married people get more affected by Type 2 ,whereas unmarried subjects were found to be more affected by Type 1. There is a significant effect of gender and marital status, and also the influence of specific food and lifestyle habits listed is evident. Conclusions The antiquity of early descriptions of diabetes underscores the importance of the observation and recording of medical conditions as humans evolve; more similar studies should be conducted to validate the data in ancient medicinal texts and to prove their authenticity in the modern scientific era.

Keywords Diabetes mellitus (DM) . . Health survey . Data collection

Introduction complications and increases the rate of mortality in diabetic patients. DM (Type 1, Type 2, and gestational) is one of the Diabetes mellitus (DM) is a serious chronic metabolic disorder diseases most commonly encountered by the healthcare pro- characterized by a hyperglycaemic state, as a result of chronic fessionals. Type 1 diabetes is caused by autoimmune destruc- insulin resistance, which leads to pancreatic beta cell dysfunc- tion of pancreatic β cells in genetically predisposed individ- tion and subsequently a massive failure in insulin secretion uals, and results in severe insulin deficiency with a requirement (dos Santos et al. 2014); and older adults with diabetes repre- for treatment with insulin. It is typically considered a disease of sent a full spectrum of health statuses ranging from excellent or childhood and adolescence, but can occur at any age. Type 2 good health to very vulnerable or frail. Diabetes is a condition diabetes is predominantly a disease of adulthood and is asso- which if left untreated can develop womb-to-tomb ciated with obesity, insulin resistance, and relative but not ab- solute insulin deficiency (Thomas et al. 2017). Gestational diabetes (GDM) is a temporary condition that occurs in preg- nancy and carries a long-term risk of type 2 diabetes. * Priyadharshan Ranganathan DM has emerged as a serious health-care problem world- [email protected] wide. Diabetes progressively contains a larger proportion of patients newly diagnosed every year (Rashedi et al. 2017). As 1 Department of Human Genetics and Molecular Biology, Bharathiar of 2015, 415 million adults have diabetes, and this number is University, Coimbatore, Tamil Nadu 641 046, India estimated to increase to 642 million by 2040. Population growth 2 Human Molecular Cytogenetics and Stem Cell Laboratory, and ageing have contributed to this increase, but are not solely Department of Human Genetics and Molecular Biology, Bharathiar University, Coimbatore, Tamil Nadu 641 046, India responsible for it. The prevalence (age-standardized) of diabetes 582 J Public Health (Berl.): From Theory to Practice (2019) 27:581–590 is growing in all regions. This is especially the case in India, quantity of sweet urine (Lakhtakia 2013). Mehanoi and with its fast-growing economy and diverse population, with prameha are the two words coined by the Siddha and varying levels of literacy, income, traditional and cultural be- system of : they are also classified liefs, and varieties of diet pattern (Tharkar et al. 2015), and into 21 types and 20 types each and description of the diabetes is one of the most prevalent chronic diseases. Over disease's signs and symptoms are clearly mentioned in 40% of children fewer than 5 years are malnourished, and it Thirumoolarkarukidainikandu, Mehanoinidhanam has also become known as a Bdiabetes capital^ of the world, (Siddha medical text book) and in Carahanidhana with an estimated 65+ million diabetic patients aged 20– (Ayurveda medical text book) respectively. The script 79 years in 2013, and substantial further increases anticipated in detail mentions the root cause of why mehanoi (Wells et al. 2016). There were over 72 million cases of diabetes (diabetes) arises, and its symptoms and causative agents in India in 2017; subsequently, this number is set to increase to are explained. The personal, social, and economic costs 592 million by 2035 (Upadhyay et al. 2013). of diabetes are huge, and are likely to adversely affect Siddha is one of the ancient medical systems in India, con- India’s economic development over the next couple of sidered as the mother medicine of ancient /Dravidians decades. Unless urgent steps are taken to thwart this in South India. The word Siddha means established truth burgeoning epidemic, more and more young and (Shukla and Saraf 2011;Piet1952). The Siddha system is a middle-aged Indians will fall prey to diabetes in the treasure house of secret science, embodying the results of the prime of their lives. ardent pursuit thereof by the ancient . This civiliza- Hence, the prime aim of the current study is to understand tion dates back to 12,000 years B.C. Two thousand years ago, the correlation and to validate the knowledge of Indian ancient the traditional medical system of the Tamils was known as manuscripts which mention types of meham and prameha, its Marunthu (medicine) (Shukla and Saraf 2011). Thousands symptoms and contributory agents that lists food as well as of examples of Siddha literature still remain in the form of lifestyle habits together as one of the reasons explained for palm-leaf manuscripts. Of late, Siddha is slowly gaining rec- DM. Therefore and additional seven new questions were cre- ognition in the world of complementary/ . ated to evaluate the credibility and judge the legitimacy of the Modern medicine recognizes Siddha medicine as an alterna- texts with respect to modern trends in food and lifestyle, and to tive East Indian medical system prevalent among Tamil- assess the possibility of their correlation. We also try at the speaking people (Stephen 2005). The Ayurvedic concept ap- same time to evaluate the ubiquity of DM and its tie-in with peared and developed between 2500 and 500 BC in India respect to age, gender, and marital status. (Subhose et al. 2005). The literal meaning of Ayurveda is Bscience of life,^ because the ancient Indian system of health care focused on views of man and his illness. It has been Materials and methods pointed out that positive health means metabolically well- balanced human beings. Ayurveda is also called the Bscience Study design of longevity^ because it offers a complete system for living a long healthy life (Pandey et al. 2013). A questionnaire-based cross-sectional and descriptive study The ancient Indian physician Sushruta and the sur- was conducted in the regions of Tamil Nadu and Kerala during geon Charaka (400–500 A.D.) were able to identify the period 2015–2017. The study areas were mainly the two types, later to be named Type I and Type II Coimbatore, Erode, Tirupur, Vellore, and some regions in diabetes. A disease characterised by the ‘too great emp- Kerala which included Palakkad, Kozhikode, and tying of urine’ finds its place in antiquity through Malappuram. The study included inhabitants aged 18–89 years Egyptian manuscripts dating back to 1500 B.C. Indian of the two states. We had screened out 1206 subjects with Type physicians called it madhumeha (‘honey urine’) because 1, Type 2, and gestational diabetes (668 men and 538 females), it attracted ants. The term prameha has two parts: ‘pra’ out of 4500 subjects that we surveyed. A questionnaire was meaning abundant and ‘meha’ meaning ‘passing of large designed which consisted of 58 descriptive questions covering quantity of urine’. Incidentally the term diabetes was their general information, diabetic knowledge, food style derived from the Greek term diabainein to mean ‘to followed before being diagnosed with diabetes, physical activ- cross through a siphon ’ meaning 'continuous free flow ity, blood sugar monitoring, and type of medication was used of water', and applied to mean elimination of large to collect the data of patients with regard to their follow-up. quantity of urine. Thus it can be seen that the terms ‘prameha’ and ‘diabetes’ are synonymous, while the Study population terms 'madhumeha' and 'diabetes mellitus’ have a simi- lar meaning: madhu and mellitus mean honey and thus The study began in 2015 when males and females 18 to madhumeha and diabetes mellitus mean passing of large 89 years of age responded to a questionnaire designed J Public Health (Berl.): From Theory to Practice (2019) 27:581–590 583 according to the prescribed format. The study sample Table 1 General characteristics of the subjects screened was randomly selected from lists of all inhabitants in Characteristics F % the age range 18–89 years (reflecting the overall focus on healthy aging). The questionnaire included an overall General information assessment of the patient’s health, medical history, life- Age (years) 16–35 428 35.4 style, and socioeconomic indicators. All questions were 36–55 573 47.51 translated from English into both Tamil and Malayalam 56–75 180 14.9 languages and back into English in order to check for 76–95 25 2.07 Gender accuracy. Subjects were consecutively recruited until Male 668 55.4 quotas for sex, age, and marital status were addressed. Female 538 44.6 Any subject who, when asked, stated that they were Marital status Unmarried 168 13.9 non-diabetic was excluded from recruitment. Subjects Married 1038 86.1 were asked to complete a survey on socioeconomic sta- Family tus, health, and lifestyle and food habits. All subjects Nuclear 891 73.9 Joint 314 26 were provided with proper knowledge about the study Socio-economic status before being analysed. Upper middle class 219 18.1 Middle class 618 51.4 Lower middle class 208 17.2 Inclusion and exclusion criteria Unanswered 161 13.3 Diabetic history The inclusion criteria for the patients who participated were: – Type of diabetes affected with? They were aged above 18 years, regardless of their caste, Type 1 288 23.9 Type 2 785 65.1 gender, and religion or socio-economic status. They had been Gestational 127 10.5 diagnosed with either type of diabetes for at least 1 year. They – Age at which you were diagnosed with? were taking diabetes medication, and they were able to speak, 10-29 448 37.14 30-59 709 58.7 read, and write either in English or their native language. 60-89 14 1.16 On the other hand, the exclusion criteria were: participants Unanswered 35 2.90 who were non-diabetic or had any conditions other than dia- Medical and health information – Do you smoke? betes, and those who were not able to read or write in English Yes 142 11.8 or their native language. No 1063 88.1 Former smoker – Do you drink alcohol? Study sampling and data collection Daily 32 2.7 Weekly 32 2.7 A 55-item study instrument that consisted of 11 sections (A, Monthly 61 5.1 Rarely 207 17.2 B, C, D, E, F, G, H, I, J, K) in total was developed for data Never 872 72.3 collection. Section A contained 11 questions that were re- Diabetes questions and knowledge lated to respondent’s general details (age, gender, current – Have you received your diabetic education before? height, current weight, occupation, and family system were Yes 1174 97.3 No 29 2.4 independent variables of the study). Section B contained Unanswered 3 0.2 two questions regarding their diabetic history. Section C – How would you rate your understanding of diabetes? contained seven questions that were designed to evaluate Very good 52 4.3 Good 857 71.1 their previous food and lifestyle before they were been di- Fair 244 20.2 agnosed with DM. Section D contained five questions relat- Poor 53 4.4 ed to medical and health conditions. Section E contained Description of physical activity Heavy 172 14.2 three questions that were used to evaluate the details regard- Moderate 273 22.6 ing pregnancy, which were to be answered only by women Light 289 23.9 respondents. Section F contained two questions to assess None 472 39.1 their diabetic knowledge. Section G contained a total of F = frequency, % = percentage ten questions that dealt with their blood sugar monitoring details (including the details of tests performed by the indi- viduals). Section H consisted of four questions about their queries regarding the medication used. Questions were de- physical wellbeing. Section I consisted of questions regard- signed in multiple choice format; some questions were mea- ing their stress. Section J contained 11 questions regarding sured using a nominal scale (yes/no). For the purpose of their food style and nutrition. Lastly, section K contained achieving optimum results, both English, Tamil, and 584 J Public Health (Berl.): From Theory to Practice (2019) 27:581–590

Table 2 Relationship within the study samples: cross-tabulation between gender and the type of diabetes affected with

Gender Type of diabetes affected with Total

Not known Gestational Type 1 Type 2

Female Count 1 128 109 300 538 % within gender 0.2% 23.8% 20.3% 55.8% 100.0% % within type of diabetes affected with 16.7% 100.0% 38.0% 38.2% 44.6% % of total 0.1% 10.6% 9.0% 24.9% 44.6% Male Count 5 0 178 485 668 % within gender 0.7% 0.0% 26.6% 72.6% 100.0% % within type of diabetes affected with 83.3% 0.0% 62.0% 61.8% 55.4% % of total 0.4% 0.0% 14.8% 40.2% 55.4% Total Count 6 128 287 785 1206 % within gender 0.5% 10.6% 23.8% 65.1% 100.0% % within type of diabetes affected with 100.0% 100.0% 100.0% 100.0% 100.0% % of total 0.5% 10.6% 23.8% 65.1% 100.0%

%=percentage

Malayalam (regional languages of Tamil Nadu and Kerala) Data analysis versions of the questionnaires were developed. The ques- tionnaire was translated into the regional languages, and All data were analysed using Microsoft Excel (Microsoft translated back into English to ensure that the essential Corporation, Redmond, WA, USA) and SPSS version 21. meaning of questionnaire remained preserved. Random checks were performed to detect errors in data entry.

Table 3 Relationship within the study samples: cross-tabulation between marital status and the type of diabetes affected with

Type of diabetes affected with Marital status Total

Married Unmarried

Unanswered Count 3 3 6 % within type of diabetes affected with 50.0% 50.0% 100.0% % within marital status 0.3% 1.8% 0.5% % of total 0.2% 0.2% 0.5% Gestational Count 127 0 127 % within type of diabetes affected with 100.0% 0.0% 100.0% % within marital status 12.2% 0.0% 10.5% % of total 10.5% 0.0% 10.5% Type 1 Count 181 107 288 % within type of diabetes affected with 62.8% 37.2% 100.0% % within marital status 17.4% 63.7% 23.9% % of total 15.0% 8.9% 23.9% Type 2 Count 727 58 785 % within type of diabetes affected with 92.6% 7.4% 100.0% % within marital status 70.0% 34.5% 65.1% % of total 60.3% 4.8% 65.1% Total Count 1038 168 1206 % within type of diabetes affected with 86.1% 13.9% 100.0% % within marital status 100.0% 100.0% 100.0% % of total 86.1% 13.9% 100.0%

%=percentage J Public Health (Berl.): From Theory to Practice (2019) 27:581–590 585

Table 4 Relationship within the study samples: cross-tabulation between gender and the consumption of following food stuffs in excess

Gender Had any of the following food stuffs in excess

Buttermilk Curd Ghee Milk None

Female Count 50 153 40 114 181 % within gender 9.3% 28.4% 7.4% 21.2% 33.6% % who had any of the following foodstuffs in excess 18.6% 67.7% 45.5% 45.1% 48.9% % of total 4.1% 12.7% 3.3% 9.5% 15.0% Male Count 219 73 48 139 189 % within gender 32.8% 10.9% 7.2% 20.8% 28.3% % who had any of the following foodstuffs in excess 81.4% 32.3% 54.5% 54.9% 51.1% % of total 18.2% 6.1% 4.0% 11.5% 15.7% Total Count 269 226 88 253 370 % within gender 22.3% 18.7% 7.3% 21.0% 30.7% % who had any of the following foodstuffs in excess 100.0% 100.0% 100.0% 100.0% 100.0% % of total 22.3% 18.7% 7.3% 21.0% 30.7%

%=percentage

Student’s t-test and the chi-square test were applied as at cross-checking the causative factors of DM as mentioned in appropriate. traditional medical texts.

Prameha and madhumeha Statistical analysis

Interestingly, a great deal of knowledge is available in tradi- We performed the data analysis using IBM® SPSS® Complex tional Indian medicine, both Siddha and Ayurveda, but still it Sample version 21. Descriptive statistics were used, while fre- is not validated due to lack of scientific research and evidence- quencies (f) and percentages (%) were examined for categor- based follow-up. In this study, the unique criteria is the set of ical variables. The chi-square statistic is used for testing rela- questions we incorporated in the questionnaire which enquire tionships between categorical variables, and the null hypothe- about the subject’s health and lifestyle conditions before they sis of the chi-square test is that no relationship exists on the were diagnosed with DM. The questions were mainly aimed categorical variables in the population; they are independent.

Table 5 Relationship within the study samples: cross-tabulation between gender and the habit of eating of too much sweet content

Gender Had an habit of eating too much sweet content Total

No Yes

Female Count 197 341 538 % within gender 36.6% 63.4% 100.0% % who had a habit of eating too much sweet content 38.5% 49.1% 44.6% % of total 16.3% 28.3% 44.6% Male Count 315 353 668 % within gender 47.2% 52.8% 100.0% % who had a habit of eating too much sweet content 61.5% 50.9% 55.4% % of total 26.1% 29.3% 55.4% Total Count 512 694 1206 % within gender 42.5% 57.5% 100.0% % who had a habit of eating too much sweet content 100.0% 100.0% 100.0% % of total 42.5% 57.5% 100.0%

%=percentage 586 J Public Health (Berl.): From Theory to Practice (2019) 27:581–590

Table 6 Relationship within the study samples: cross-tabulation between gender and regular consumption of meat with high fat content

Gender Consumed meat with high fat regularly Total

No Yes

Female Count 454 84 538 % within gender 84.4% 15.6% 100.0% % who consumed meat with high fat regularly 49.6% 29.0% 44.6% % of total 37.6% 7.0% 44.6% Male Count 462 206 668 % within gender 69.2% 30.8% 100.0% % who consumed meat with high fat regularly 50.4% 71.0% 55.4% % of total 38.3% 17.1% 55.4% Total Count 916 290 1206 % within gender 76.0% 24.0% 100.0% % who consumed meat with high fat regularly 100.0% 100.0% 100.0% % of total 76.0% 24.0% 100.0%

%=percentage

Results type 2, followed by 109 (26.6%) affected with type 1 and one (0.24%) who did not respond to this question. Table 1 depicts the general characteristics of the subjects Table 3 shows that of total sample size of 1206 subjects, screened with diabetes. 1038 (86.1%) were married and 168 (13.9%) were unmarried. Table 2 shows that the highest number of subjects (785: Of the married category total of 1038, 727 (70%) were affect- 65.1%) were affected with type 2, compared to type 1 with ed with type II, 181 (17.4%) with type I, and 127 (12.2%) with 287 (23.8%) followed by gestation diabetes with 128 (10.6%) gestational diabetes. Of the unmarried subjects total of 168, and six (0.5%) unanswered. Among a total of 668 male can- 107 (63.7%) were affected with type I, and 58 (34.5%) with didates, 485 (72.6%) were affected with type 2 diabetes and type II. 178 (26.6%) with type 1; five respondents (0.7%) did not Table 4 shows that of 668 male subjects in the total answer the question. Among a total of 538 female candidates, study group, 479 (71.7%) had consumed one or more of if the 118 affected with gestational diabetes are excluded and the food items mentioned in section C to excess, and we consider 410 as 100%, 73.0% (300) were affected with 189 (28.3%) of the group had not taken any of the food

Table 7 Relationship within the study samples: cross-tabulation between gender and consumption of water with high salt content

Gender Consumed water with salt content borewell water Total

No Yes

Female Count 335 203 538 % within gender 62.3% 37.7% 100.0% % who consumed water with salt content borewell water 38.8% 59.2% 44.6% % of total 27.8% 16.8% 44.6% Male Count 528 140 668 % within gender 79.0% 21.0% 100.0% % who consumed water with salt content borewell water 61.2% 40.8% 55.4% % of total 43.8% 11.6% 55.4% Total Count 863 343 1206 % within gender 71.6% 28.4% 100.0% % who consumed water with salt content borewell water 100.0% 100.0% 100.0% % of total 71.6% 28.4% 100.0%

%=percentage J Public Health (Berl.): From Theory to Practice (2019) 27:581–590 587

Table 8 Relationship within the study samples: cross-tabulation between gender and habit of eating late

Gender Ever had an habit of eating late Total No Yes Female Count 242 296 538 % within gender 45.0% 55.0% 100.0% % who ever had an habit of eating late 39.1% 50.4% 44.6% % of total 20.1% 24.5% 44.6% Male Count 377 291 668 % within gender 56.4% 43.6% 100.0% % who ever had an habit of eating late 60.9% 49.6% 55.4% % of total 31.3% 24.1% 55.4% Total Count 619 587 1206 % within gender 51.3% 48.7% 100.0% % who ever had an habit of eating late 100.0% 100.0% 100.0% % of total 51.3% 48.7% 100.0% Chi-square tests Value Df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Pearson chi-square 15.654a 1 .000 Continuity correctionb 15.199 1 .000 Likelihood ratio 15.683 1 .000 Fisher’s Exact test .000 .000 N of valid cases 1206

%=percentage a 0 cells (.0%) have expected count less than 5. The minimum expected count is 261.86 b Computed only for a 2 × 2 table Df, degrees of freedom; Asymp., asymptotic; Sig., significance items mentioned. Of 538 female subjects in the total above, the highest consumed food stuff was buttermilk study group, 357 (66.3%) had consumed one or more with 219 (45.7%), and the least was ghee with 48 of the food items mentioned in section C to excess, and (10.0%). Of the 357 female consumers, the highest con- 181 (33.6%) of the group had not taken any of the food sumed foodstuff was curd with 153 (42.8%) and the items mentioned. Among the 479 male consumers least was ghee with 40 (11.0%).

Table 9 Relationship within the study samples: cross-tabulation between gender and habit of walking too much as a daily routine

Gender Too much walking as a daily routine Total

No Yes

Female Count 460 78 538 % within gender 85.5% 14.5% 100.0% % who do too much walking as a daily routine 50.4% 26.5% 44.6% % of total 38.1% 6.5% 44.6% Male Count 452 216 668 % within gender 67.7% 32.3% 100.0% % who do too much walking as a daily routine 49.6% 73.5% 55.4% % of total 37.5% 17.9% 55.4% Total Count 912 294 1206 % within gender 75.6% 24.4% 100.0% % who do too much walking as a daily routine 100.0% 100.0% 100.0% % of total 75.6% 24.4% 100.0%

%=percentage 588 J Public Health (Berl.): From Theory to Practice (2019) 27:581–590

Table 10 Relationship within the study samples: cross-tabulation between gender and habit of sleeping late at night

Gender Habit of sleeping late at night Total

No Yes

Female Count 145 393 538 % within gender 27.0% 73.0% 100.0% % with habit of sleeping late at night 31.2% 53.0% 44.6% % of total 12.0% 32.6% 44.6% Male Count 320 348 668 % within gender 47.9% 52.1% 100.0% % with habit of sleeping late at night 68.8% 47.0% 55.4% % of total 26.5% 28.9% 55.4% Total Count 465 741 1206 % within gender 38.6% 61.4% 100.0% % with habit of sleeping late at night 100.0% 100.0% 100.0% % of total 38.6% 61.4% 100.0%

%=percentage

Table 5 shows that of the total sample size of 1206, 694 socio-political history, and immense diversity of culture, (57.5%) responded ‘yes’ and 512 (42.5%) responded ‘no’. dialects, and customs, and it has varying levels of liter- Table 6 shows that 916 (76.0%) of people responded ‘no’ acy, income, traditional and cultural beliefs, and varieties for consumption of meat with high fat content and 290 of diet pattern. (24.0%) responded ‘yes’. In our study, we examined the population-based associa- Table 7 shows that of the total subjects surveyed, 863 tions between gender, age, marital status, and the prevalence (71.6%) responded ‘no’ to Bconsuming water with high salt of DM in some of the districts of Tamil Nadu and Kerala. The content^ and 343 (28.4%) responded ‘yes’. study reported significant gender difference in the prevalence Table 8 shows that 619 (51.3%) of people responded ‘yes’ of DM, which is supported by evidence from other studies, to the question Bhabit of eating late^ and 587 (48.7%) which show a male preponderance (Anjana et al. 2011; responded ‘no’. Meshram et al. 2016; Tesfaye et al. 2016), although a few Table 9 shows that of the total subjects screened, 912 studies have reported no gender differences (Ramachandran (75.6%) responded ‘no’ and 294 (24.4%) responded ‘yes’ to et al. 2001; Barik et al. 2016; Goswami et al. 2016; Tripathy the question Bhabit of walking too much as a daily routine^. et al. 2017). Some studies found a relationship between DM Table 10 shows that 741 (61.4%) responded ‘yes’ and 465 and gender (Lidfeldt et al. 2007; Robbins et al. 2001); how- (38.6%) responded ‘no’ to the question Bhabit of sleeping late ever, data from other studies agree with our finding (Azimi- at night^. Nezhad et al. 2008;Bosietal.2009). From the results we obtained, it is clear that irrespective of age and gender, people were aware of knowing the risk of Discussion and conclusion consuming high-fat food, and that is visible in the values for consuming ghee as the least preferred, followed by curd, milk, Diabetes mellitus is one of the most prominent non- and buttermilk. But as per the traditional Siddha texts, con- communicable diseases that are undermining the health suming too much of any of this four items gives a high risk of of the people of India and placing additional burdens on acquiring DM, and that is evident in our study. The lifestyle health systems (Hilawe et al. 2013). India is the diabetes changes also give us an idea about the authenticity of the result capital in terms of percentage, as home to 69.1 million of our study. One of the most consistent findings in the Indian people with DM, the second highest number of cases epidemiology studies dealing with diabetes is that the preva- after China. Recent epidemiological evidence indicates lence of Type 2 diabetes increases with age (Zargar et al. 2000; a rising DM epidemic across all classes, both the affluent Ramachandran 2005;Ramachandranetal.1988, 1997, 2001; and the poor in India (Tripathy et al. 2017). Diabetes is Mohan et al. 2001;Iyeretal.2001;Ashaetal.2001; Sadikot also one of the most prevalent chronic diseases in India. et al. 2004). Studies conducted in different countries other India is a gigantic, heterogeneous country with an ap- than India also give similar data which showed a significant proximate population of 1.1 billion people, a complex difference in the prevalence of DM with respect to increasing J Public Health (Berl.): From Theory to Practice (2019) 27:581–590 589 age (Azimi-Nezhad et al. 2008;Dray-Spiraetal.2008; data. Lifestyle exposure data was collected by face-to-face Rahmanian et al. 2013). interview, with an extensive review of the range of food and Our study is limited by the cross-sectional nature of the data, lifestyle choices followed by the subjects before they were which does not provide any indication of the direction of effect affected by DM, with an intention to understand which unique or causality. This limitation also prevents any measure of tem- habits were common among all the subjects. The stress factor poral changes in prevalence of DM and factors associated with was self-reported by the patients themselves, and the severity DM. Longitudinal studies would complement the present study of stress was studied. To best of our knowledge, this is the first to determine causality and directional effect of the factors. In study of this type which extracts ideas from the traditional addition, in this study we have tried to incorporate both tradi- Siddha and Ayurveda systems in DM and trying to validate tional as well as scientific knowledge on diabetes, and have them with modern-life parameters, taking them into consider- attempted to create a correlated study with the incorporation ation for further study to find out the causes of DM, and trying of certain novel questions which according to the traditional to understand the modus operandi of how DM originates. Siddha medical text are mentioned among the various causes of this particular condition. As far as we are aware, the present Acknowledgements We would like to thank the authorities of Bharathiar trial is the first study which has tried to utilise both science and University, Coimbatore for providing infrastructure facilities for this re- search work. We are grateful to the authors of the articles included in the traditional medical knowledge in the case of diabetes mellitus. reference, many of whom kindly provided us with additional information In this study, there was a significant difference observed in regarding their studies. We also express our gratitude to Bioline the prevalence of DM between the married and the unmarried Diagnostics and Labs, Coimbatore for their support in completion of group, whereas in other studies no significant difference has the research work. Last but not least, we thank the patients and their families for their dedicated participation in completion of the work. been reported (Azimi-Nezhad et al. 2008; Rahmanian et al. This research did not receive any specific grant from funding agencies 2013; Tesfaye et al. 2016). Even though association was in the public, commercial, or not-for-profit sectors. established between diabetes mellitus and physical activity, it is difficult to prove the causal relationship as the number Compliance with ethical standards of physically active subjects with the disease was very low. However, it is evident that many cross-sectional (WHO 2008; Ethical approval All adopted procedures and methodology used were in Abu-Aisha et al. 2008; Enang et al. 2014; Olatunbosun et al. agreement with the ethical standards for research at the university. 1998) and cohort studies (Willi et al. 2007;Wangetal.2013; Conflict of interest All authors declare that they have no conflict of Jee et al. 2010; Kamaura et al. 2011; Baliunas et al. 2009)have interest. documented lack of physical activity as an established factor for having diabetes mellitus. The first step to combating any epidemic is to gather all available information about it. Until References recently, efforts to tackle the diabetes epidemic in India were hamstrung by the lack of quality original data emanating from Abu-Aisha H, Elhassan EA, Khamis AH, Abu-Elmaali A (2008) the country itself. It was also evident to researchers in this field Hypertension and obesity in police forces households in that the data obtained from Western populations could not be Khartoum, Sudan: a pilot report-part of the Bpolice forces hyperten- directly extrapolated to Indians, since the natural history of sion, diabetes, renal insufficiency, and thyroid derangements (HyDRIT) study^, Sudan. Sudanese J Public Health 3:17–25 diabetes in Indians is quite different from other ethnic groups. Anjana RM, Pradeepa R, Deepa M, Datta M, Sudha V, Unnikrishnan R, The antiquity of early descriptions of diabetes underscores Bhansali A, Joshi SR, Joshi PP, Yajnik CS, Dhandhania VK (2011) the importance of the observation and recording of medical Prevalence of diabetes and prediabetes (impaired fasting glucose conditions as humans evolve. Early physicians used whatever and/or impaired glucose tolerance) in urban and rural India: phase – was there in their pursuit of knowledge, skills, and diagnosis. I results of the Indian Council of Medical Research INdiaDIABetes (ICMR–INDIAB) study. Diabetologia 54:3022–3027 From unrecorded accounts to published knowledge, this hu- Asha PB, Murthy BN, Chellamariappan M, Gupte MD, Krishnaswami man scourge is, simply put, a modern-day epidemic. We, and CV (2001) Prevalence of known diabetes in Chennai City. J Assoc future generations from the field of life science and medicine, Physicians India 49:974–981 share the task of taking this history forward. Additional sim- Azimi-Nezhad M, Ghayour-Mobarhan MP, Parizadeh MR, Safarian M, ilar studies should be conducted to validate the information in Esmaeili H, Parizadeh SM, Khodaee G, Hosseini J, Abasalti Z, Hassankhani B, Ferns G (2008) Prevalence of type 2 diabetes ancient medicinal texts and to prove its authenticity in the mellitus in Iran and its relationship with gender, urbanisation, edu- modern scientific era. cation, marital status and occupation. Singap Med J 49:571 In our case–control study, the subjects were selected ran- Baliunas DO, Taylor BJ, Irving H, Roerecke M, Patra J, Mohapatra S, domly with respect to age, gender, and socio-economic and Rehm J (2009) Alcohol as a risk factor for type 2 diabetes: a sys- tematic review and meta-analysis. Diabetes Care 32:2123–2132 work conditions. Educational level may be related to a better Barik A, Mazumdar S, Chowdhury A, Rai RK (2016) Physiological and working and living environment with exposure to awareness behavioral risk factors of type 2 diabetes mellitus in rural India. BMJ about the problem, hence selection bias is a possibility in our Open Diabetes Res Care 4:e000255 590 J Public Health (Berl.): From Theory to Practice (2019) 27:581–590

Bosi PL, Carvalho AM, Contrera D, Casale G, Pereira MA, Gronner MF, Ramachandran A, Snehalatha C, Kapur A, Vijay V, Mohan V, Das AK, Diogo TM, Torquarto MT, Oishi J, Leal AM (2009) Prevalence of Rao PV, Yajnik CS, Kumar KP, Nair JD (2001) Diabetes diabetes and impaired glucose tolerance in the urban population of Epidemiology Study Group in India (DESI). High prevalence of 30 to 79 years of the city of São Carlos, São Paulo. Arq Bras diabetes and impaired glucose tolerance in India: National Urban Endocrinol Metabol 53:726–732 Diabetes Survey. Diabetologia 44:1094–1101 dos Santos PF, dos Santos PR, Ferrari GS, Fonseca GA, Ferrari CK Rashedi V, Asadi-Lari M, Delbari A, Fadayevatan R, Borhaninejad V, (2014) Knowledge of diabetes mellitus: does gender make a differ- Foroughan M (2017) Prevalence of diabetes type 2 in older adults: ence? Osong Public Health Res Perspect 5:199–203 findings from a large population-based survey in Tehran, Iran (urban Dray-Spira R, Gary TL, Brancati FL (2008) Socioeconomic position and HEART-2). Diabetes Metab Syndr Clin Res Rev 11:S347–S350 cardiovascular disease in adults with and without diabetes: United Robbins JM, Vaccarino V, Zhang H, Kasl SV (2001) Socioeconomic States trends, 1997–2005. J Gen Intern Med 23:1634 status and type 2 diabetes in African American and non-Hispanic Enang OE, Otu AA, Essien OE, Okpara H, Fasanmade OA, Ohwovoriole white women and men: evidence from the third National Health and AE, Searle J (2014) Prevalence of dysglycemia in Calabar: a cross- Nutrition Examination Survey. Am J Public Health 91:76 sectional observational study among residents of Calabar, Nigeria. Sadikot SM, Nigam A, Das S, Bajaj S, Zargar AH, Prasannakumar KM, BMJ Open Diabetes Res Care 2:e000032 Sosale A, Munichoodappa C, Seshiah V, Singh SK, Jamal A (2004) Goswami AK, Gupta SK, Kalaivani M, Nongkynrih B, Pandav CS The burden of diabetes and impaired fasting glucose in India using (2016) Burden of hypertension and diabetes among urban popula- the ADA 1997 criteria: prevalence of diabetes in India study ≥ tion aged 60 years in South Delhi: a community based study. J Clin (PODIS). Diabetes Res Clin Pract 66:293–300 Diagn Res 10:LC01 Shukla S, Saraf S (2011) Fundamental aspect and basic concept of siddha Hilawe EH, Yatsuya H, Kawaguchi L, Aoyama A (2013) Differences by medicines. Sys Rev Pharm 2(1):48 sex in the prevalence of diabetes mellitus, impaired fasting glycaemia Stephen ES (2005) Complementary and alternative medicine. In: Dennis and impaired glucose tolerance in sub-Saharan Africa: a systematic LK, Eugene B, Anthony SF, Stephen LH, Llarry JL (eds) Harrison’s review and meta-analysis. Bull World Health Organ 91:671–82D principles of internal medicine, 16th edn. McGraw–Hill Medical, Iyer SR, Iyer RR, Upasani SV, Baitule MN (2001) Diabetes mellitus in New York, p 67 Dombivli—an urban population study. J Assoc Physicians India 49: 713–716 Subhose V, Srinivas P, Narayana A (2005) Basic principles of pharma- ceutical science in Ayurvĕda. Bull Indian Inst Hist Med Hyderabad Jee SH, Foong AW, Hur NW, Samet JM (2010) Smoking and risk for – diabetes incidence and mortality in Korean men and women. 35(2):83 92 Diabetes Care 33:2567–2572 Tesfaye T, Shikur B, Shimels T, Firdu N (2016) Prevalence and factors Kamaura M, Fujii H, Mizushima S, Tochikubo O (2011) Weight gain and associated with diabetes mellitus and impaired fasting glucose level risk of impaired fasting glucose after smoking cessation. J among members of federal police commission residing in Addis Epidemiol 21:431–439 Ababa, Ethiopia. BMC Endocr Disord 16:68 Lakhtakia R (2013) The history of diabetes mellitus. Sultan Qaboos Univ Tharkar S, Devarajan A, Barman H, Mahesh U, Viswanathan V (2015) Med J 13:368 How far has translation of research been implemented into clinical Lidfeldt J, Li TY, Hu FB, Manson JE, Kawachi I (2007) A prospective practice in India? Are the recommended guidelines adhered to? Int J study of childhood and adult socioeconomic status and incidence of Diabetes Mellit 3:25–30 type 2 diabetes in women. Am J Epidemiol 165:882–889 Thomas NJ, Jones SE, Weedon MN, Shields BM, Oram RA, Hattersley Meshram II, Rao MV, Rao VS, Laxmaiah A, Polasa K (2016) Regional AT (2017) Frequency and phenotype of type 1 diabetes in the first variation in the prevalence of overweight/obesity, hypertension and six decades of life: a cross-sectional, genetically stratified survival diabetes and their correlates among the adult rural population in analysis from UK Biobank. Lancet Diabetes Endocrinol 6(2):122– India. Br J Nutr 115:1265–1272 129 Mohan V, Shanthirani S, Deepa R, Premalatha G, Sastry NG, Saroja R Tripathy JP, Thakur JS, Jeet G, Chawla S, Jain S, Pal A, Prasad R, Saran R (2001) Intra urban differences in the prevalence of the metabolic (2017) Prevalence and risk factors of diabetes in a large community- syndrome in southern India—theChennaiUrbanPopulationStudy based study in North India: results from a STEPS survey in Punjab, (CUPS no. 4). Diabet Med 18:280–287 India. Diabetol Metab Syndr 9:8 Olatunbosun ST, Ojo PO, Fineberg NS, Bella AF (1998) Prevalence of Upadhyay RP,Misra P, Chellaiyan VG, Das TK, Adhikary M, Chinnakali diabetes mellitus and impaired glucose tolerance in a group of urban P, Yadav K, Sinha S (2013) Burden of diabetes mellitus and predi- adults in Nigeria. J Natl Med Assoc 90:293 abetes in tribal population of India: a systematic review. Diabetes Pandey MM, Rastogi S, Rawat AK (2013) Indian traditional ayurvedic Res Clin Pract 102:1–7 system of medicine and nutritional supplementation. Evid Based Wang Y,Ji J, Liu YJ, Deng X, He QQ (2013) Passive smoking and risk of Complement Alternat Med 2013:376327 type 2 diabetes: a meta-analysis of prospective cohort studies. PLoS Piet JH. (1952) A logical presentation of the Śaiva Siddhānta Philosophy. One 8:e69915 Doctoral dissertation, Christian Literature Society for India (United Wells JC, Pomeroy E, Walimbe SR, Popkin BM, Yajnik CS (2016) The Society for Christian Literature) elevated susceptibility to diabetes in India: an evolutionary perspec- Rahmanian K, Shojaei M, Jahromi AS (2013) Relation of type 2 diabetes tive. Front Public Health 4:145 mellitus with gender, education, and marital status in an Iranian Willi C, Bodenmann P, Ghali WA, Faris PD, Cornuz J (2007) Active urban population. Rep Biochem Mol Biol 1:64 — smoking and the risk of type 2 diabetes: a systematic review and Ramachandran A (2005) Epidemiology of diabetes in India three de- meta-analysis. Jama 298:2654–2664 cades of research. J Assoc Physicians India 53:34–38 World Health Organization (2008) Waist circumference and waist–hip Ramachandran A, Jali MV, Mohan V, Snehalatha C, Viswanathan M ratio: report of a WHO expert consultation. WHO, Geneva, pp 8–11 (1988) High prevalence of diabetes in an urban population in South India. Bmj 3:587–590 Zargar AH, Khan AK, Masoodi SR, Laway BA, Wani AI, Bashir MI, Dar Ramachandran A, Snehalatha C, Latha E, Vijay V, Viswanathan M FA (2000) Prevalence of type 2 diabetes mellitus and impaired glu- cose tolerance in the Kashmir Valley of the Indian subcontinent. (1997) Rising prevalence of NIDDM in an urban population in – India. Diabetologia 40:232–237 Diabetes Res Clin Pract 47:135 146